Sign in with
Sign up | Sign in
Your question

Aquamark world record broken with Conroe System

Last response: in CPUs
Share
April 25, 2006 12:24:43 PM

2.72GHz Conroe and overclocked x1900 crossfire. All air, stock cooling, nothing exotic (it'll be interesting to see what happens when they move on to more extreme overclocking).

Preemptive: Yes I know that Aquamark is largely a graphics driven benchmark. The actual score isn't even the point. The point is that as the Conroe engineering samples are being hooked up, world records are falling left and right. That is truly exciting, it is a great time for competition and enthusiasts in the computer industry.

Preemptive: This is not fake, free cable guy has years of experience obtaining ES chips from Intel and AMD and doing amazing overclocking with both brands. No ulterior motives, just a technology enthusiast.

Link: Conroe/x1900 Xfire Aquamark

Cheers!
April 25, 2006 12:57:45 PM

8O 8O 8O This is on air cooling?! 8O 8O 8O
April 25, 2006 1:11:12 PM

nice, I expected to be better. anyway I am waiting for MMM and 9-inch. What do they have to cry about the slow FX overclocked at 3.7GHz with dry ice :lol: 
April 25, 2006 1:11:26 PM

Yup pure air cooling, nothing exotic. In the thread I linked FCG has provided the screenshots of the previous record holder, an FX-60 on phase cooling, overclocked to 3.7GHz and a pair of massively overclocked 7900GTXs.

The Conroe chip is 2.13GHz stock and is still only running at 2.72GHz and is only air cooled. I'm looking forward to seeing the effect as the Conroe chip is ramped even higher.

Cheers!
April 25, 2006 1:43:10 PM

This is amazing!!By next year they would be selling regular desktop like Dell or Gateway at a high end price that could outperforms FX-60 and 840 EE systems! This is insane! 8O Now I'm gonna sell my chip for $300 and get this new Conroe! My pc is going to explode! :twisted:
April 25, 2006 1:48:35 PM

hmmm....too bad Conroe isnt released yet.
Another "engineering sample"....
April 25, 2006 2:14:32 PM

Bravo <clapping hands> if this is the power of a 4 issue intel core then amd and k10 should be a behemoth! :D 
April 25, 2006 2:51:54 PM

Quote:
2.72GHz Conroe and overclocked x1900 crossfire. All air, stock cooling, nothing exotic (it'll be interesting to see what happens when they move on to more extreme overclocking).

Preemptive: Yes I know that Aquamark is largely a graphics driven benchmark. The actual score isn't even the point. The point is that as the Conroe engineering samples are being hooked up, world records are falling left and right. That is truly exciting, it is a great time for competition and enthusiasts in the computer industry.

Preemptive: This is not fake, free cable guy has years of experience obtaining ES chips from Intel and AMD and doing amazing overclocking with both brands. No ulterior motives, just a technology enthusiast.

Link: Conroe/x1900 Xfire Aquamark

Cheers!




This is even more reason why $1000 965 EEs WILL NOT SELL. Intel is going to take a real bath because of this chip. It's unfortunate the P4 sucked so bad but to go from that to Conroe means NO ONE will want P4s now. Good job, Intel.
April 25, 2006 2:58:41 PM

World records are falling left and right? What other records?

I'm glad intel finally got off their ass and did something right :p  too bad for them it was too late to keep my business
April 25, 2006 3:20:14 PM

Quote:
2.72GHz Conroe and overclocked x1900 crossfire. All air, stock cooling, nothing exotic (it'll be interesting to see what happens when they move on to more extreme overclocking).

Preemptive: Yes I know that Aquamark is largely a graphics driven benchmark. The actual score isn't even the point. The point is that as the Conroe engineering samples are being hooked up, world records are falling left and right. That is truly exciting, it is a great time for competition and enthusiasts in the computer industry.

Preemptive: This is not fake, free cable guy has years of experience obtaining ES chips from Intel and AMD and doing amazing overclocking with both brands. No ulterior motives, just a technology enthusiast.

Link: Conroe/x1900 Xfire Aquamark

Cheers!




This is even more reason why $1000 965 EEs WILL NOT SELL. Intel is going to take a real bath because of this chip. It's unfortunate the P4 sucked so bad but to go from that to Conroe means NO ONE will want P4s now. Good job, Intel.

This thread has nothing to do with 965 EEs. You are a strange fellow, arguing that Intel having a great chip that outperforms everything is somehow BAD for Intel and not AMD. Please stay on topic in this thread.
April 25, 2006 3:22:02 PM

Quote:
World records are falling left and right? What other records?

I'm glad intel finally got off their ass and did something right :p  too bad for them it was too late to keep my business

Aquamark, SuperPi and 3dMark05 so far. Give folks time to run more benchmarks, it looks promising.

Regardless of which company is doing this, it is always fun to see the state of the art improving.

Cheers!
April 25, 2006 3:29:58 PM

uhh, and how much can we expect conroes to sell for?
April 25, 2006 3:40:44 PM

Quote:
uhh, and how much can we expect conroes to sell for?

This is a bit off topic, but the roadmap pricing is as follows:

E6700: 2.67GHz, 4MB L2, 1066FSB, $529
E6600: 2.40GHz, 4MB L2, 1066FSB, $310
E6400: 2.13GHz, 2MB L2, $240
E6300: 1.86GHz, 2MB L2, $210.
April 25, 2006 3:45:30 PM

Quote:
uhh, and how much can we expect conroes to sell for?

This is a bit off topic, but the roadmap pricing is as follows:

E6700: 2.67GHz, 4MB L2, 1066FSB, $529
E6600: 2.40GHz, 4MB L2, 1066FSB, $310
E6400: 2.13GHz, 2MB L2, $240
E6300: 1.86GHz, 2MB L2, $210.

yup, was just going to post this.

also:
http://www.theinquirer.net/?article=31126
April 25, 2006 3:50:54 PM

Quote:
nice, I expected to be better. anyway I am waiting for MMM and 9-inch. What do they have to cry about the slow FX overclocked at 3.7GHz with dry ice

:roll:

It would be nice to have benchmarks that included profile information so that the quality of the benchmark could be determined. Does this benchmark show actual performance? Unless there is a lot more to this benchmark than I'm aware of the answer is easily, no.

Just based on common sense for writing a benchmark like this I know that the core graphic engine will have sections of code that are hit most of the time. It is a big question whether these core sections can remain in the L2 cache. If the code that is running most of the time stays in the cache this will have a big effect on speed. Obviously too this benchmark will have to access a graphic database that tells it how to draw the scene so memory bandwidth will be a factor. Almost all of the calculations will be done in floating point, presumably with SSE so SSE speed will be a factor.

So, at a minimum, cache size, cache speed, memory bandwidth, and SSE peformance will be factors. However, there are other factors as well like branch predictability.

It would be nice if people could understand concepts like tuning and performance averaging. Basically, if you varied the parameters used in the graphic engine and graphed this you could by changing each item finally arrive at code that achieved a maximum level of performance for a given set of hardware with a given set of data. Now, if you go back and examine these graphs and you discover that your tuned configuration has a sharp narrow peak then all of your efforts have been wasted. Why? Because a sharp narrow peak means that the tuning will only work in a narrow range and will fall off as things change. This is the oppposite of robust code which may be a bit slower but will handle changes without a sharp deterioration in performance. Now, expanding this means that not only should a graphic engine give high average performance with varying parameters but it should also give high average performance on varying hardware. If some of you still don't understand what I'm talking about let me give you an example.

Suppose I profiled code for a given graphic engine and optimized the code so that key parts would fit inside the L1 cache. Suppose this then gave a big boost in performance to AMD processors and noticeably less on Intel's because of the much smaller L1. Would this be a real world example of graphic performance? The answer is, no. Why? Because if I were really writing a graphic engine I would want it to have the broadest possible performance which would include both AMD and Intel hardware. I wouldn't use tuned versions unless I had one for each. For example, I could check the cpu ID and then change code sections depending on which processor was found.

Without knowing the average level of performance for this code on varying data and hardware there is no way to determine the quality of the score. Roughly, it should indicate increases in performance within a given hardware family but that is something you can typically tell without running any benchmark.

Absent what is needed to determine the quality of aquamark the only alternative is to sample a lot of benchmarks and get a picture of overall performance. I realize that I have a programming background but Is this concept really so hard to understand?
April 25, 2006 4:35:37 PM

Hi,
Do you know if any news on the K10?
Thanks.
April 25, 2006 4:57:34 PM

Quote:
2.72GHz Conroe and overclocked x1900 crossfire. All air, stock cooling, nothing exotic (it'll be interesting to see what happens when they move on to more extreme overclocking).

Preemptive: Yes I know that Aquamark is largely a graphics driven benchmark. The actual score isn't even the point. The point is that as the Conroe engineering samples are being hooked up, world records are falling left and right. That is truly exciting, it is a great time for competition and enthusiasts in the computer industry.

Preemptive: This is not fake, free cable guy has years of experience obtaining ES chips from Intel and AMD and doing amazing overclocking with both brands. No ulterior motives, just a technology enthusiast.

Link: Conroe/x1900 Xfire Aquamark

Cheers!




This is even more reason why $1000 965 EEs WILL NOT SELL. Intel is going to take a real bath because of this chip. It's unfortunate the P4 sucked so bad but to go from that to Conroe means NO ONE will want P4s now. Good job, Intel.

This thread has nothing to do with 965 EEs. You are a strange fellow, arguing that Intel having a great chip that outperforms everything is somehow BAD for Intel and not AMD. Please stay on topic in this thread.


I'm sorry I don't drool over benchmarks as you do but my point was that releasing all this now when they are so fragile on Wall Street will kill even more profits. I love that they finally don't look like Cyrix, but maybe they should have let the Core Duo and Solo come out this year and waited until next year for Conroe. What happens if AMD K8L beats Conroe ( which is a possibilty)?
Then Intel will have no high margin desktop chips and a dwindling server share. Not good fo the bottom line. I mention this because Intel did everyhing they could to destroy AMD (some maybe illegally) and now their profits are down ASP is down and now they still have several months before this new wonder chip which will make P4s worth even less, so they have the choice of shifting ALL manuf to Conroe and taking a bath on P4s or shifting some and having an overstock of chips no one wants.

I guess sometimes speed does kill!
April 25, 2006 5:03:52 PM

Quote:
2.72GHz Conroe and overclocked x1900 crossfire. All air, stock cooling, nothing exotic (it'll be interesting to see what happens when they move on to more extreme overclocking).

Preemptive: Yes I know that Aquamark is largely a graphics driven benchmark. The actual score isn't even the point. The point is that as the Conroe engineering samples are being hooked up, world records are falling left and right. That is truly exciting, it is a great time for competition and enthusiasts in the computer industry.

Preemptive: This is not fake, free cable guy has years of experience obtaining ES chips from Intel and AMD and doing amazing overclocking with both brands. No ulterior motives, just a technology enthusiast.

Link: Conroe/x1900 Xfire Aquamark

Cheers!
8O Kick ass!
April 25, 2006 5:16:34 PM

It just makes me wonder what the scores would be with one of those Vapo Chill things ? I mean would they be able to get a 60% OC ? I think those super cooling things are one way to see where a process will top out at in terms of Mhz :) 
April 25, 2006 5:34:17 PM

Quote:
2.72GHz Conroe and overclocked x1900 crossfire. All air, stock cooling, nothing exotic (it'll be interesting to see what happens when they move on to more extreme overclocking).

Preemptive: Yes I know that Aquamark is largely a graphics driven benchmark. The actual score isn't even the point. The point is that as the Conroe engineering samples are being hooked up, world records are falling left and right. That is truly exciting, it is a great time for competition and enthusiasts in the computer industry.

Preemptive: This is not fake, free cable guy has years of experience obtaining ES chips from Intel and AMD and doing amazing overclocking with both brands. No ulterior motives, just a technology enthusiast.

Link: Conroe/x1900 Xfire Aquamark

Cheers!




This is even more reason why $1000 965 EEs WILL NOT SELL. Intel is going to take a real bath because of this chip. It's unfortunate the P4 sucked so bad but to go from that to Conroe means NO ONE will want P4s now. Good job, Intel.

This thread has nothing to do with 965 EEs. You are a strange fellow, arguing that Intel having a great chip that outperforms everything is somehow BAD for Intel and not AMD. Please stay on topic in this thread.

I think it has some viability on the post and is well on topic. Price to performance has become a leading indicator of performance (quoted from both AMD and Intel) as is heat and power consumption. If pure performance was the only reason for this thread then Intel and AMD both make specialized chips which would smoke anything we could get our hands on.

In the last few years instead of going forward both companies step back as they teach their CPUs new tricks. To me it feels like a con, stepping back in Mhz every time we move ‘forward’. Well, long story short, EE will remain viable because of this stepping back. Sure, they will drop in price, but they will still be offered and most likely at prices that will still tempt the bargain shopper. Intel will loose nothing because their fabrication plants will stop producing them, slowly phasing them out for the newer better fabs.
April 25, 2006 5:40:00 PM

Quote:
2.72GHz Conroe and overclocked x1900 crossfire. All air, stock cooling, nothing exotic (it'll be interesting to see what happens when they move on to more extreme overclocking).

Preemptive: Yes I know that Aquamark is largely a graphics driven benchmark. The actual score isn't even the point. The point is that as the Conroe engineering samples are being hooked up, world records are falling left and right. That is truly exciting, it is a great time for competition and enthusiasts in the computer industry.

Preemptive: This is not fake, free cable guy has years of experience obtaining ES chips from Intel and AMD and doing amazing overclocking with both brands. No ulterior motives, just a technology enthusiast.

Link: Conroe/x1900 Xfire Aquamark

Cheers!


I am not weighing in on Conroes performance one way or the other, though I will say that I still haven't seen a completely credible review of its actual performance, and won't until it is released.

However, my question is this: How many "I love Conroe, it is so sweet even though it hasn't even been released yet!" threads can we possibly have in this forum?
April 25, 2006 5:57:41 PM

Quote:
2.72GHz Conroe and overclocked x1900 crossfire. All air, stock cooling, nothing exotic (it'll be interesting to see what happens when they move on to more extreme overclocking).

Preemptive: Yes I know that Aquamark is largely a graphics driven benchmark. The actual score isn't even the point. The point is that as the Conroe engineering samples are being hooked up, world records are falling left and right. That is truly exciting, it is a great time for competition and enthusiasts in the computer industry.

Preemptive: This is not fake, free cable guy has years of experience obtaining ES chips from Intel and AMD and doing amazing overclocking with both brands. No ulterior motives, just a technology enthusiast.

Link: Conroe/x1900 Xfire Aquamark

Cheers!

:o :o  YAWN, YAWN, YAWN :o :o 
So what's clocked at 2.72GHz, the FSB?? Jam tomorrow, bullsh1t today!!!!
Let's wait till it ships!!
April 25, 2006 6:12:34 PM

People are funny. I thought this was an enthusiast board. I guess people just want to sit with their head int he sand and not pay attention to any information until "it ships".

This is merely info, take it or leave it. But he fact remains, this system now holds the world record for Aquamark3. And that is interesting, regardless of the conclusions you want to draw from it.
April 25, 2006 6:17:20 PM

It holds a record??? HOw??
Its not even been released. They have an "engineering sample", thats it. Nobody has seen ANY retail conroe benchmarks. Period.
I bet you were impressed with Prescotts benchmarks when it came out too right. How wrong you were.
April 25, 2006 6:20:54 PM

Quote:
It holds a record??? HOw??

By obtaining the highest score in the benchmark of any system on record, ever.

If a prerelease AM2 system get's a higher score, then it has the record. How are you having trouble understanding that? Bizarre.
April 25, 2006 6:30:11 PM

Quote:
2.72GHz Conroe and overclocked x1900 crossfire. All air, stock cooling, nothing exotic (it'll be interesting to see what happens when they move on to more extreme overclocking).

Preemptive: Yes I know that Aquamark is largely a graphics driven benchmark. The actual score isn't even the point. The point is that as the Conroe engineering samples are being hooked up, world records are falling left and right. That is truly exciting, it is a great time for competition and enthusiasts in the computer industry.

Preemptive: This is not fake, free cable guy has years of experience obtaining ES chips from Intel and AMD and doing amazing overclocking with both brands. No ulterior motives, just a technology enthusiast.

Link: Conroe/x1900 Xfire Aquamark

Cheers!


"Legitimate (adj.) - conforming to acknowledged standards: complying with recognized rules, standards, or traditions"

There is no official standard since there are no official Conroe processors out there. It may be right, or maybe the official conroe CPUs could be better--but it is not legitimate. I doubt the guy's a "fake" or anything either.

I'm only sick of reviewing benchmark tests of non-existing hardware. They cannot be confirmed as there is no "real" conroe to stack it up against. Maybe the Intel "engineers" designed a special instruction mod that allows it to excel with Aquamark software.

Test the conroe sample with 3dMark, PCMark, CPU-z, SiSoft Sandra, Winstone, and any other benchmark I forgot about...maybe I might lean towards believing it. Even then--we won't know for certain until it is officially released.
April 25, 2006 6:37:18 PM

Quote:
2.72GHz Conroe and overclocked x1900 crossfire. All air, stock cooling, nothing exotic (it'll be interesting to see what happens when they move on to more extreme overclocking).

Preemptive: Yes I know that Aquamark is largely a graphics driven benchmark. The actual score isn't even the point. The point is that as the Conroe engineering samples are being hooked up, world records are falling left and right. That is truly exciting, it is a great time for competition and enthusiasts in the computer industry.

Preemptive: This is not fake, free cable guy has years of experience obtaining ES chips from Intel and AMD and doing amazing overclocking with both brands. No ulterior motives, just a technology enthusiast.

Link: Conroe/x1900 Xfire Aquamark

Cheers!


"Legitimate (adj.) - conforming to acknowledged standards: complying with recognized rules, standards, or traditions"

There is no official standard since there are no official Conroe processors out there. It may be right, or maybe the official conroe CPUs could be better--but it is not legitimate. I doubt the guy's a "fake" or anything either.

I'm only sick of reviewing benchmark tests of non-existing hardware. They cannot be confirmed as there is no "real" conroe to stack it up against. Maybe the Intel "engineers" designed a special instruction mod that allows it to excel with Aquamark software.

Test the conroe sample with 3dMark, PCMark, CPU-z, SiSoft Sandra, Winstone, and any other benchmark I forgot about...maybe I might lean towards believing it. Even then--we won't know for certain until it is officially released.

Indeed, this is but one data point, no one should draw any further conclusions from it. Over time, hopefully data for all of the Benchmarks you list will become available (though CPU-z is not a benchmark). It will be interesting to say the least.

Implying that there are special instructions in the chip specifically designed to excel in Aquamark is pretty childish though.
April 25, 2006 6:38:16 PM

Quote:
People are funny. I thought this was an enthusiast board. I guess people just want to sit with their head int he sand and not pay attention to any information until "it ships".

This is merely info, take it or leave it. But he fact remains, this system now holds the world record for Aquamark3. And that is interesting, regardless of the conclusions you want to draw from it.


I guess I would be more impressed if it wasn't a DX9 benchmark that gets more gains from the Dual x1900s than from the CPU. I am very much an enthusiast, however, I am also hesitant to make a judgement until I see some real comparison numbers that are actually comparing CPUs, and not systems.

I am impressed by Intels moves with Conroe, frankly, its the first intelligent desktop chip design decision they've made in a few years. However, I don't have a short memory, and have seen many many pre-release claims using "engineering samples", (which are usually better than the release version) that have proven to be less than accurate when it comes to actual retail products. Therefoe, I prefer to wait until I see real benchmarks to make a judgement.
April 25, 2006 6:41:55 PM

Indeed, I agree, no one should be coming anywhere close to passing "judgement" on any of these chips. This is just interesting information. And over time more and more will become available, and then eventually conclusions will be drawn. However, that doesn't mean we should avoid information as it becomes available. That is what enthusiast boards are all about.

Cheers!
April 25, 2006 6:49:22 PM

It's a next gen architecture. Shouldn't it beat all the records?

The intel fanboys must be starved for any news....
April 25, 2006 6:51:36 PM

Very true, I have seen too many people drawing unwarranted conclusions from single points of information about conroe.

Even more so in the server space, which is OT, but I haven't seen anything from Intel that looks even close to catching AMD there.
April 25, 2006 6:54:11 PM

Quote:
Implying that there are special instructions in the chip specifically designed to excel in Aquamark is pretty childish though.


Yeah, it was :roll: . Anyway, my main point is regarding the fanboyism stuff:

"Conroe is better than Athlon 64 FX processors" --Probably correct...however, this statement is designed to create childish debates over unofficial and preliminary information.

"Preliminary tests show that Conroe will best the current Athlon 64 FX processors." --While immature folks will still criticise this, it is no longer "stretching" the truth--hence, this is the best way to remark.

Treat facts as facts and opinions, possibilities, and probabilities for what they are. Quickest way to start a fanboy thread is by making a statement without it being a proven or accepted truth. :arrow:

(What the he** am I putting this here for? Preaching to the choir thus far...)
April 25, 2006 7:01:47 PM

Quote:
2.72GHz Conroe and overclocked x1900 crossfire. All air, stock cooling, nothing exotic (it'll be interesting to see what happens when they move on to more extreme overclocking).

Preemptive: Yes I know that Aquamark is largely a graphics driven benchmark. The actual score isn't even the point. The point is that as the Conroe engineering samples are being hooked up, world records are falling left and right. That is truly exciting, it is a great time for competition and enthusiasts in the computer industry.

Preemptive: This is not fake, free cable guy has years of experience obtaining ES chips from Intel and AMD and doing amazing overclocking with both brands. No ulterior motives, just a technology enthusiast.

Link: Conroe/x1900 Xfire Aquamark
Who cares about Aquamark3? The program is so outdated, no one should take it seriously; next you're going to show us some Quake2 benchmarks.
April 25, 2006 7:13:28 PM

bourgeoisdude:

Come on, all of iterations' threads are generally inflammatory touting of results of an overclocked platform without any specifications that could be verified and / or repeated, and are fanboyish in nature. So I don't think you're preaching to a choir at all.

This is all but a silly debate. Once final hardware comes out, we will be able to run appropriate comparisons. Until then, all is speculation. Besides, I don't know if running tests with OC'ed parts proves a point. If you want to say that Conroe is great, why would you use an OC'ed GFX card?
April 25, 2006 7:21:32 PM

Lame. I'm only interested in the cutting edge results of the technology we follow here. I'm all for fair comparisons of parts. I'm not for jumping to conclusions. I'm all for making info available, and I don't think that is inflammatory. The discussion becomes inflamatory when haters of either brand pile on and start insulting each other and drawing inaccurate conclusions from very limited data.

Regardless of what company makes the chip, when a long standing benchmark record is broken, I find it interesting.
April 25, 2006 7:29:39 PM

Quote:
nice, I expected to be better. anyway I am waiting for MMM and 9-inch. What do they have to cry about the slow FX overclocked at 3.7GHz with dry ice

:roll:

It would be nice to have benchmarks that included profile information so that the quality of the benchmark could be determined. Does this benchmark show actual performance? Unless there is a lot more to this benchmark than I'm aware of the answer is easily, no.
................Absent what is needed to determine the quality of aquamark the only alternative is to sample a lot of benchmarks and get a picture of overall performance. I realize that I have a programming background but Is this concept really so hard to understand?
1 benchmark is not enough to determinate performance. Anyway aquamark is benchmarking system performance for various jobs and is giving a scalable objective results.
I do programm and I do understand how CPUs are working. In this case the cache capacity and cache speed are not the main factor. Yes the performance is dependend on them, but much more is dependend on the architecture of the CPU. Core will be the best mainstream architecture when its chips will be available on the market. There are some intelectualy limited uneducated discutants(fanboys can be educated, they are not) that can not accept that, and I am just provocating them. I found making no-sence discussions interesting on forumz like this one :lol: 
April 25, 2006 7:33:07 PM

Do you really not realize that posting a thread titled "World record broked with Conroe" is implying that Conroe is the best thing out there? Do you really not make that connection?

And then people read that OC'ed graphics cards were used, etc. etc. What purpose does it really serve then if not to invite flame wars. Is it an official result? No. Was it done in a manner that may be reproduced by somebody else? Big no. People are complaining left and right that even the settings utilized for the bench are not available. Not to mention the parts are not a) widely available; b) more importantly, final production samples.

You're just using the scores to allow you to post these threads. What is the purpose? It is not even good information.
April 25, 2006 7:39:13 PM

The only thing that the title implies is that this Conroe system broke the world record for Aquamark, which it did.

Of course there are OC'd graphics cards involved. The previous record holder was a Phase Cooled AMD FX-60 overclocked to 3.7GHz with overclocked graphics cards as well. It is not meant to be reproduced, it is a world record. If everyone could reproduce it, that would be pointless.

You seem to think that a world record run is somehow something people should take into account when purchasing a system. That is awfully silly.
April 25, 2006 7:50:22 PM

But overclocking the videocards to a certain speed makes it questionable. If the CPU was already so fast, why did he bother to overclock the GPUs?

Also, the idea of a 2.7GHz Conroe beating a 3.7GHz FX60 is insane; I smell something fishy. Don't get me wrong, the Conroe will be an awesome CPU, I just don't believe it'll outperform the FX60 by this much.
April 25, 2006 7:58:36 PM

Quote:
Also, the idea of a 2.7GHz Conroe beating a 3.7GHz FX60 is insane; I smell something fishy. Don't get me wrong, the Conroe will be an awesome CPU, I just don't believe it'll outperform the FX60 by this much.

And why not! A 2.6Ghz FX beat a 3.8Ghz P4???

At this stage of the game an engineering sample that has been lent/given to an enthusiast to overclock will pretty much be a final release candidate. The only special bit about it, is that it will overclock well.
The internal logic/circuit will not be any different to the end product that we will get (maybe a few bug fixes).
April 25, 2006 8:03:10 PM

Quote:
And why not! A 2.6Ghz FX beat a 3.8Ghz P4???
We're comparing a 12 stage K8 to a 14 stage Conroe; this is why I'm not seeing how a Conroe can beat an FX60 by that much.
April 25, 2006 8:08:29 PM

This is primarily a Directx9 benchmark, so the CPU ratings have a lot less to do with it than the GPUs. I can see either beating the other, since this really isn't in any way a CPU benchmark, let alone comparison.
April 25, 2006 8:37:42 PM

HOLY CRAP. AND THIS WAS WITH A STOCK INTEL BOARD, WHICH ARE NOTORIOUSLY CRAPPY OC'ERS. On LN2/ANY phase changer, there's a good chance this could be a 4ghz beast...
April 25, 2006 8:44:47 PM

Hmmm, sorry guys, but it's just an engineering sample. If we get these results when they are actually released, that's when I'll believe and really care.
April 25, 2006 8:46:58 PM

Quote:
Hmmm, sorry guys, but it's just an engineering sample. If we get these results when they are actually released, that's when I'll believe and really care.

I highly doubt this was in any way a hoax. Every claimed review has shown the same type of thing. Remember when Prescott came out? Nobody really had any reviews or records collapsing that hinted at good performance. This should be much different.
April 25, 2006 9:41:33 PM

Quote:
Hmmm, sorry guys, but it's just an engineering sample. If we get these results when they are actually released, that's when I'll believe and really care.

I agree with you, but I believe the results just becouse I have tested Dothan and Yonah. Conroe is imporved Yonah with some nice technologies, instrcutions sets, better x86 decoder, better branch prediction, more execution units and 128bit ex.units. If I consider them in my calc, yeap I belive there will be no apropriate chip that will beat Conroe at the price it will cost.
April 26, 2006 12:02:22 AM

Quote:
But overclocking the videocards to a certain speed makes it questionable. If the CPU was already so fast, why did he bother to overclock the GPUs?

Also, the idea of a 2.7GHz Conroe beating a 3.7GHz FX60 is insane; I smell something fishy. Don't get me wrong, the Conroe will be an awesome CPU, I just don't believe it'll outperform the FX60 by this much.


:roll: Too right-there IS something fishy going on!!!!!!

Remember Intel's NetBurst architecture? Amongst other things, it had a
"Rapid Execution Engine", a term used to describe the fact that the ALUs
were clocked at 2X the core frequency!!!!!! So, a 1.4GHz Pentium 4 had ALUs running ar 2.8GHz!! Has the penny dropped??????Maybe this is what they are doing with CONroe?? I note that in all their propaganda about their
"Core Architecture", not a lot is said about frequency-viz, which bits of the CPU work at what frequency.
April 26, 2006 12:42:54 AM

I'm impressed by the work the guy put into it. The numbers look nice n all but I am going to have to see many more benchmarks showing at least similar scores.

Besides I have an Athlon XP 2600 and I don't wanna have to scrap my mobo.

*Is an AMD fanboy*
April 26, 2006 1:12:20 AM

Quote:
I'm impressed by the work the guy put into it. The numbers look nice n all but I am going to have to see many more benchmarks showing at least similar scores.

Besides I have an Athlon XP 2600 and I don't wanna have to scrap my mobo.

*Is an AMD fanboy*

Lol, Conroe doesn't release on Socket A :D 
April 26, 2006 3:39:42 AM

Quote:
But overclocking the videocards to a certain speed makes it questionable. If the CPU was already so fast, why did he bother to overclock the GPUs?

Also, the idea of a 2.7GHz Conroe beating a 3.7GHz FX60 is insane; I smell something fishy. Don't get me wrong, the Conroe will be an awesome CPU, I just don't believe it'll outperform the FX60 by this much.


:roll: Too right-there IS something fishy going on!!!!!!

Remember Intel's NetBurst architecture? Amongst other things, it had a
"Rapid Execution Engine", a term used to describe the fact that the ALUs
were clocked at 2X the core frequency!!!!!! So, a 1.4GHz Pentium 4 had ALUs running ar 2.8GHz!! Has the penny dropped??????Maybe this is what they are doing with CONroe?? I note that in all their propaganda about their
"Core Architecture", not a lot is said about frequency-viz, which bits of the CPU work at what frequency.

you gusy are lacking something...

the Pentium M at a lower clock beats the X2 at a higher clock. So Intel has already shown it has better performance, Mhz wise. It's not hard for me to believe a 2.7Ghz Conroe beating a FX-60 at 3.7Ghz. After all, remember? Didn't a AMD Athlon 64 at 2.0Ghz beat a P4 3.0Ghz? :roll:
Things have just reversed this time around, IMO.
!